The BARD clinical trial aims to determine the most appropriate add-on therapy for Black individuals who are not responding to low dose inhaled corticosteroids (ICS). This unique cohort offers repeated sampling across 4 treatment regimens and provides an opportunity to compare response in children and adults. Using samples obtained at baseline and following 4 treatment regimens, this ancillary study proposes to significantly augment clinical and genetic data from BARD with mechanistic information through quantitation of Th2, inflammatory, and resolving markers and through the use of complementary metabolomics techniques. We hypothesize that ICS-responsive individuals will have a Th2-high phenotype, reflected in increased IL4, IL13, IL5, IgE, and periostin. We have previously found a relationship between urinary leukotriene E4 (uLTE4) and asthma, and between sphingolipids and COPD. Therefore we will test the hypothesis that LTE4 and other pro-inflammatory molecules are increased in ICS- responsive individuals. Because asthma is a heterogeneous condition, knowing the dynamics of multiple molecules will be required in order to fully understand mechanism, and we will utilize an unbiased metabolomics approach to discover new markers. We expect that response to medication can be predicted based on the relative abundance of Th2, inflammatory, and other disease-related molecules and that levels will revert to normal levels in response to medication. Finally, we expect that groups of individuals will cluster based on mechanistic differences. Therefore, in specific aim 1 (SA1), we will quantitate Th2, inflammatory, and resolving markers and will perform metabolomics on baseline plasma, sputum (adults only), and urine samples. In SA2 we will develop prediction models, perform cluster analysis, and correlate results from sputum, plasma, and urine. In SA3 we will use longitudinally collected urine samples to determine the effect of increased ICS therapy on specific molecules. This high impact clinical study will provide an unprecedented opportunity to determine if plasma markers can be used to predict response to therapy, thereby answering urgent questions in clinical medicine.